Kinetic Study of the Activation of Banana JuiceEnzymatic Browning by the Addition of

Maltosyl-�-cyclodextrin

JOSÉ M. LÓPEZ-NICOLÁS,*,† ANTONIO J. PÉREZ-LÓPEZ,‡

ÁNGEL CARBONELL-BARRACHINA,§ AND FRANCISCO GARCÍA-CARMONA†

Department of Biochemistry and Molecular Biology-A, Faculty of Biology, University of Murcia,Campus de Espinardo, 30071 Murcia, Spain, Department of Food Science and Technology, SanAntonio Catholic University, Avda. Los Jerónimos s/n, 30107 Guadalupe (Murcia), Spain, and

Departamento de Tecnología Agroalimentaria, Universidad Miguel Hernández, Carretera de Beniel,km. 3,2, 03312 Orihuela (Alicante), Spain

In recent years, the use of cyclodextrins (CDs) as antibrowning agents in fruit juices has receivedgrowning attention. However, there has been no detailed study of the behavior of these moleculesas substances, which can lead to the darkening of foods. In this paper, when the color of fresh bananajuice was evaluated in the presence of different CDs, the evolution of several color parameters wasthe opposite of that observed in other fruit juices. Moreover, a kinetic model based on the complexationby CDs of the natural browning inhibitors present in banana is developed for the first time to clarifythe enzymatic browning activation of banana juice. Finally, the apparent complexation constantbetween the natural polyphenoloxidase inhibitors present in banana juice and maltosyl-�-CD wascalculated (Kci ) 27.026 ( 0.212 mM-1).

KEYWORDS: Cyclodextrin; browning; banana; juice; color

INTRODUCTION

In the last decade, the consumption of tropical fruit juices inSpain has increased as consumers look for new foods withdifferent sensorial properties (1, 2). One of the most consumedtropical fruits in Spain is banana, with its enormous bananaworld production (3, 4), important nutritional value (5, 6), andwidely appreciated flavor and aroma (7, 8) contributing to thispopularity. However, the evolution of banana juice color duringprocessing and storage is one of the principal inconveniencesof this tropical juice, because color degradation impairs thesensory properties of the product and also lowers its nutritionalvalue.

One of the main factors that can alter the color of bananajuice and therefore limit its commercial shelf life, is browningbecause of the enzymatic activity of polyphenol oxidase (PPO)(monophenol dihydroxyphenylalanine: oxygen oxidoreductases,EC 1.14.18.1) (9, 10). The presence of PPO in banana fruit hasbeen widely reported (10–15), and several authors in the lastdecade have focused on the use of postharvest chemicaltreatments to avoid enzymatic browning through the inhibition

of PPO (16–25). Moreover, some authors have studied thepossible interaction between the oxidation products of enzymaticbrowning (i.e., quinones) and other compounds present in themedium (26–28).

In recent decades, there has been growing interest in usingantioxidants in food (29). However, although some of the mostwidely used antioxidants are very effective as antibrowningagents, they are synthetic products and tend to feed the growingconcern about the potential health hazards of synthetic sub-stances (30). Consumers are demanding a reduction in theoverall use of chemicals in fresh products, and therefore,alternative methodologies need to be investigated (31). In recentyears, the use of cyclodextrins (CDs) as natural antibrowningagents has been looked at for many foods (32–36). Recently, apaper concerning the use of CDs to slow the enzymaticbrowning of apple juice was published by our group and akinetic model to clarify the controversy existing in the literatureabout the mechanism through which enzymatic browning isinhibited by CDs was proposed (37).

However, although these natural antioxidant agents can beused to prevent the enzymatic browning of fresh juice fruit, adetailed study of its effects on the sensorial properties of manyjuices must be developed. Indeed, many new antioxidantcompounds proposed as antibrowning agents may well act aspromotors of enzymatic browning depending upon such factorsas the concentration, type of food to be protected, antioxidantsource, etc. One example of this type of behavior concerns the

* To whom correspondence should be addressed: Department ofBiochemistry and Molecular Biology-A, Faculty of Biology, Universityof Murcia, Campus de Espinardo, 30071 Murcia, Spain. Fax: 34-968-364765. E-mail: josemln@um.es.

† University of Murcia.‡ San Antonio Catholic University.§ Universidad Miguel Hernández.

J. Agric. Food Chem. 2007, 55, 9655–9662 9655

10.1021/jf0713399 CCC: $37.00 2007 American Chemical SocietyPublished on Web 10/12/2007

use of CDs. Although CDs are generally recognized as anti-browning agents, in a previous work, we studied the role ofCDs as a modulator of latent banana pulp PPO and a browningactivating effect of CDs was described (38). However, no kineticmodel was proposed to explain the results, and the effect ofthe addition of CDs on CIELAB color parameters of bananajuices was not studied. Moreover, further studies were requiredto compile a complete range of fruits or raw materials for whichbrowning is activated or inhibited by CDs. In the present paper,we study the effect of CD on the browning of the mostconsumed fruit juices from different sources in Spain.

Bearing in mind that the activating effect of CD on bananaenzymating browning has not been studied in detail, the effectof adding of CDs on CIELAB color parameters of banana juicehas been evaluated. For this, several factors, which may affectthe color evolution of banana juice, such as the type andconcentration of CDs, have been studied. Moreover, to clarifythe mechanism of banana juice enzymatic browning inhibitionby CD, a kinetic model has been proposed. Finally, the apparentcomplexation constants for the complexes between the naturalinhibitors of the enzymatic browning present in banana juiceand different types of CD have been calculated.

MATERIAL AND METHODS

Materials. R- and �-CD were purchased from Sigma-Aldrich(Madrid, Spain) and used as received. Maltosyl-�-CD was kindlysupplied by Ensuiko Sugar Refining Co. Ltd. (Japan). AnhydrousD-glucose was supplied by Prolabo (Fontenay-Sous-Bois, France).

Juice Preparation. Bananas (Musa acuminata subgr. Cavendish cv.Spanish Pequeña Enana), pears (Pyrus communis cv. Barlett), peaches(Prunus persina cv. Baby gold), grapes (Vitis Vinifera cv. Dominga),and apples (Malus domestica cv. Fuji) were purchased from localsupermarkets and stored at 4 °C until needed. They were peeled, cored,and sliced prior to juicing in a Moulinex Y36 blender. The fruit juicesobtained were immediately collected and mixed in a beaker containing25 mL of distilled water alone or containing enough R-CD, �-CD,maltosyl-�-CD (0–90 mM), or glucose (60 and 180 mM) to producethe final concentration of each compound indicated in eachexperiment.

Color Evolution Assessment. Color determinations were made, at25 °C, using a ColorFlex version 1.72 colorimeter (Hunterlab, Reston,VA) certified by International Organization for Standardization (ISO)9001 with a D75 light source and the observer at 10°.

The color parameters corresponding to the uniform CIELAB colorspace (39) were obtained directly from the apparatus. Within thisuniform space, two color coordinates, a* and b*, as well as apsychometric index of lightness, L*, are defined. While a* takes positivevalues for reddish colors and negative values for the greenish ones, b*takes positive values for yellowish colors and negative values for thebluish ones. L* is an approximate measurement of luminosity, whichis the property according to which each color can be considered amember on the gray scale, between black and white, within the rangeof 0–100. The total color difference (∆E*), a single value which takes

into account the differences between L*, a*, and b*, of the sampleand standard was also studied. Chroma (C*) is the attribute that allowsfor the determination of the degree of difference in comparison to agray color with the same lightness for each hue and, therefore, isconsidered the quantitative attribute of “colorful”. Hue angle (H*) isthe attribute according to which colors have been traditionally definedas reddish, greenish, etc. This is the attribute that allows a color to bedistinguished with reference to a gray color with the same lightness.This attribute is related to the differences in absorbance at differentwavelengths and is considered the qualitative attribute of color.

For all of the experiments, the previously described mixtures wereused in the color evolution assays, using the measurements at time 0as the standard. This time corresponded to the first measurement, whichwas made 1 min after the fruits had been juiced and the chemical agentsdissolved in the juice. All of the measurements were made at differenttimes during the first hour after the materials had been dissolved in thefruit juice, i.e., just when the enzymatic browning was beginning.

Three readings were obtained for each replicate to obtain uniformcolor measurements: hue (H*), total color difference (∆E*), and chroma(C*), which were calculated using equations

H*) tan–1(b*a*)

∆E*) [(∆L*)2 + (∆a*)2 + (∆b*)2]1/2

C*) [(a*)2 + (b*)2]1/2

RESULTS AND DISCUSSION

Effect of the Addition of Several Additives on the Evolu-tion of Fruit Juice Color. The effect of different additives onthe total color difference (∆E*) was measured in the differentjuices 30 min after juicing (Table 1). As expected, the additionof increasing concentrations of R-CD produced a reduction inthe ∆E* values of peach, apple, and pear juice with respect tothe values obtained in the absence of any agent, as wasdemonstrated in previous works (37, 40). However, the additionof the same concentrations of R-CD concentrations to bananajuice increased the ∆E* values compared to the control. Besidesthe fact that no study of banana juice color in the presence ofCDs had been made previously, this observation made itnecessary to carry out experiments to confirm that the colori-metric method developed in this paper is suitable for studyingthe effect of CDs on different fruit juices.

To study the possibility that inclusion complexes are formedbetween D-glucose (a molecule included in the CD structure)and the compounds responsible for enzymatic browning in fruitjuices and to confirm that the effect of CDs on color evolutionwas not due to their glucidic nature, the effect of another sugar,in this case D-glucose, on the color of apple, peach, pear, andbanana juices was studied. For this purpose, various amountsof D-glucose (60 and 180 mM), corresponding to 10 and 30mM of R-CD with regard to the number of glucose units, wereadded to the reaction medium and the CIELAB parameters weremeasured. As shown in Table 1, the ∆E* values of the differentfruit juices were similar in the absence and presence of 60 and180 mM D-glucose, although these values differed substantiallyfrom those obtained in the presence of 10 and 30 mM R-CD.These results are in good agreement with several studies thathave demonstrated that D-glucose is unable to form inclusioncomplexes with guest molecules (41, 42).

Several conclusions can be extrapolated from these data.First, CDs affect the evolution of the total color differencein the juices from many sources. Moreover, the ability ofCDs to change the color parameters of these juices is notdue to its glucidic nature but the formation of an inclusioncomplex, because no glucose/phenolic compound complexes

Table 1. Effect of Different Additives on the Evolution of Total ColorDifference (∆E*) of Several Fruit Juices from Different Sources Measured30 min after Juicing

total color difference (∆E*)

additive peach juice apple juice pear juice banana juice

none 26.01 ( 1.76 42.65 ( 1.12 28.01 ( 0.62 17.01 ( 0.94D-glucose

(60 mM)26.32 ( 1.45 41.98 ( 1.38 28.34 ( 1.31 17.10 ( 1.35

D-glucose(180 mM)

26.71 ( 1.56 42.24 ( 1.45 28.92 ( 1.87 16.95 ( 1.46

R-CD (10 mM) 20.43 ( 1.29 30.63 ( 1.38 24.11 ( 1.19 18.41 ( 0.82R-CD (30 mM) 15.09 ( 0.84 25.73 ( 1.57 17.03 ( 0.81 20.97 ( 1.24

9656 J. Agric. Food Chem., Vol. 55, No. 23, 2007 López-Nicolás et al.

exist. Finally, colorimetric studies appear to be a satisfactorymethod for observing and characterizing phenolic com-pound–CD inclusion complexes.

Comparison of the Effect of CD on the Browning ofDifferent Fruit Juices. To evaluate the effect of the additionof CDs on the enzymatic browning of fruit juices from differentsources and to compile a range of fruit juices for whichbrowning is activated or inhibited by CDs, as recommended inour previous work (38), a modified CD, in this case maltosyl-CD, was added to the reaction medium. To do this, the relativemeasurement between the light reflected and absorbed by thesamples, i.e., the lightness (L*) values of peach, pear, apple,grape, and banana juices, was measured in the absence andpresence of 90 mM maltosyl-�-CD.

The extent to which the addition of maltosyl-�-CD inhibitedenzymatic browning was expressed on a percent basis, i.e., thepercent difference between the control and treatment ∆L values1 h after juicing

% variation)∆Lcontrol -∆Ltreatment

∆Lcontrol× 100

where ∆L is the difference between the L* value at time t withthe units in minutes. As shown in Figure 1, the negative valuesof the percent variation of pear, peach, grape, and apple juicesindicate that the addition of 90 mM maltosyl-�-CD as abrowning inhibitor is effective to the extent calculated. However,the positive percent variation value obtained for banana juiceindicates the activation of the enzymatic browning by theaddition of maltosyl-�-CD.

Although several researchers are mostly interested in long-term browning control, Cheng and Crisosto (43) showed that83% of the browning measured at the end of long incubationtimes in some foods had occurred during the first hour, whichis why we have studied the behavior of banana juice enzymaticbrowning at the very beginning of the reaction (i.e., during thefirst hour). Moreover, the data presented in this paper show thatthe evolution of the total color difference (∆E*) in banana juiceat 25 °C in the absence of CD was very significant in the firsthour.

Because enzymatic browning, with a few exceptions (medlars,prunes, black raisins, dates, black figs, zapote, tea, coffee, andcocoa), is considered as product degradation, which lowers fruitquality both visually and with regard to taste and nutritionalcharacteristics, several papers have studied how CDs slowenzymatic browning in several foods (32, 33, 37, 38, 40).However, although the mechanism of fresh apple juice enzy-matic browning in the presence of maltosyl-�-CD was describedby our group (37), this is the first time that an increase in afruit juice L* value because of the addition of CD has beendemonstrated. For this reason and with the objective of clarifyingthe effect of CDs on banana juice, a detail study of the effectof different CDs on banana juice color was deemednecessary.

Effect of the CD Structure on the Browning of BananaJuice. To examine the variation of banana juice enzymaticbrowning by different types of CD, two types of natural CDgenerally recognized as safe (R-CD and �-CD) and one modifiedCD (maltosyl-�-CD) were used.

Figure 2 shows that the percent variation of banana juiceenzymatic browning is dependent upon the CD type used. Ascan be seen, the optimum variation rate was produced bymaltosyl-�-CD, followed by R-CD, while �-CD presented thelowest variation of banana juice enzymatic browning.

These results are in good agreement with those obtained inprevious studies, where maltosyl-�-CD presented the optimumstructure to modify the enzymatic browning in both apple andgrape juices (37, 40). In recent years, the use of differentfunctional groups (such as methyl, hydroxypropyl, maltosyl,acetyl, etc.) to modify natural CDs to complex many guestmolecules has been improved because many different chemicalmoieties may be introduced into the CD molecule by the reactionwith the hydroxyl groups lining the upper and lower ridges ofthe toroid (41). Moreover, the solubility of these modified CDsis generally higher than that of the respective underivatized CD.

In the case of �-CD, the low variation of banana juiceenzymatic browning may have been be due to the lowerconcentration used (10 mM), because this is the least solubleof the three CDs used in this work. Although �-CD is the mostused CD type to complex different guest molecules (44, 45),its use in the food industry is limited by its poor solubility inaqueous medium (46). Moreover, the fact that �-CD presentslower enzymatic browning variation percent values than mal-tosyl-�-CD may also be due to the stronger interactionsestablished between this type of modified CD and the guestmolecules. In a comparison of the derivatives from the samenatural CD (� derivatives), Veiga et al. (47) showed that the

Figure 1. Percent variation of enzymatic browning in several juices inthe presence of 90 mM maltosyl-�-CD. Each data point is the mean ofthree replicates.

Figure 2. Percent variation of enzymatic browning in banana juice in thepresence of different types of CDs at 90 mM. Each data point is themean of three replicates.

Activation of Banana Juice Enzymatic Browning J. Agric. Food Chem., Vol. 55, No. 23, 2007 9657

maltosyl derivative had a higher capacity to remove 1% of water,followed by the hydroxypropyl derivative and, finally, methyland acetyl derivatives. These differences could be, in general,due to the chemical structure of the substitution group, whichallows for the formation of a hydrogen bond with the hydra-tion–water molecules, because the maltosyl substitution groupgives rise to more intense bonding.

Finally, if we compare the results obtained for �-CD withthose obtained for R-CD, both natural and generally recognizedas safe, it seems that the optimum structure to activate bananajuice enzymatic browning is that formed by six molecules ofglucose, i.e., R-CD. This is probably due to the size of the CDcavity, with the inner diameter of the hydrophobic cavity beingapproximately 0.47–0.53 and 0.60–0.65 nm for R-CD and �-CD,

respectively (44). However, the variation percent of R-CD waslower than that observed for maltosyl-�-CD (Figure 2).

For these reasons, we selected maltosyl-�-CD to study theeffect of CD addition on the evolution of banana juice colorparameters and to propose a kinetic model to explain theactivation mechanism of banana juice enzymatic browning inthe presence of CDs.

Effect of the Addition of Maltosyl-�-CD on Color Evolu-tion of Banana Juice. Because there are no data in the literatureconcerning the effect of the addition of CDs on the colorparameters of banana juice, the lightness (L*), the angularcoordinates (H* and C*), and the total color difference (∆E*)were considered to define the color of this fruit juice in thepresence of these compounds.

In previous works, it was observed that the addition ofantioxidant agents, such as CDs, to fruits juice, such as appleand grape, substantially slowed and diminished the decreaseobserved for lightness (L*) in the absence of CDs (37, 40). Inthis paper and to clarify the effect of adding maltosyl-�-CD onthe evolution of banana juice lightness (L*) during the firstminutes of processing, increasing concentrations of this CD wereused (0–90 mM). As shown in Figure 3, L* fell rapidly in thefirst 60 min without CD in the medium because the juice becamedarker. However and contrary to that observed for other fruitjuices, when increasing concentrations of maltosyl-�-CD wereused, the L* value of the color scale fell ever more strongly.

The behavior of the angular coordinate H* for banana juicein the presence of CD was also the opposite of that publishedfor apple and grape juices (37, 40) (Table 2). In the case ofbanana juice, when increasing concentrations of maltosyl-�-CD were used, the slight increase in H* observed in the absenceof any agent in the first hour after juicing was converted to astrong decay of the same parameter. Finally, the chroma value(C*), a parameter that indicates the degree of saturation of colorand is proportional to the strength of the color, behaveddifferently from the H* angle and was very similar to thatobserved in the absence of any agent (Table 2).

Finally, the total color difference, ∆E*, a colorimetricparameter extensively used to characterize the variation of colorsin foods during processing, was studied in the presence ofincreasing concentrations (0–90 mM) of maltosyl-�-CD (Figure4). As can be seen, contrary to that observed for other fruitjuices studied previously, the addition of maltosyl-�-CD did

Figure 3. Effect of maltosyl-�-CD on the evolution of lightness (L*) inbanana juice at 25 °C in the absence of any CD (b) and in the presenceof maltosyl-�-CD: 10 mM (9), 30 mM ([), 60 mM (2), and 90 mM (1).Each data point is the mean of three replicates.

Table 2. Evolution of C* and H* Coordinates of Banana Juice in theAbsence and Presence of Different Maltosyl-�-CD Concentrations

Hunter valuesreactiontime (min) H* C*

without R-CD0 78.67 ( 3.21 34.98 ( 2.3420 82.92 ( 3.45 24.35 ( 2.4340 83.67 ( 3..62 17.78 ( 2.3660 82.03 ( 3.73 12.05 ( 2.64

10 mM Maltosyl-�-CD0 78.67 ( 3.21 34.98 ( 2.3420 80.26 ( 3.91 23.65 ( 2.4540 80.90 ( 3.85 18.24 ( 2.6460 80.46 ( 3.79 15.51 ( 2.62

30 mM Maltosyl-�-CD0 78.67 ( 3.21 34.98 ( 2.3420 77.16 ( 3.39 21.59 ( 2.2640 74.91 ( 3.23 15.39 ( 2.2760 74.94 ( 3.63 13.36 ( 2.45

60 mM Maltosyl-�-CD0 78.67 ( 3.21 34.98 ( 2.8620 76.06 ( 3.46 21.72 ( 2.8440 75.31 ( 3.85 19.34 ( 2.3560 74.39 ( 3.60 15.60 ( 2.21

90 mM Maltosyl-�-CD0 78.67 ( 3.21 34.98 ( 2.1220 74.05 ( 3.78 23.87 ( 2.3540 72.03 ( 3.32 19.4 ( 2.5360 70.38 ( 3.46 16.59 ( 2.63

Figure 4. Effect of the maltosyl-�-CD concentration on the evolution of thetotal color difference (∆E*) in banana juice at 25 °C in the absence of maltosyl-�-CD (b) and in the presence of maltosyl-�-CD: 10 mM (9), 30 mM ([),and 90 mM (1). Each data point is the mean of three replicates.

9658 J. Agric. Food Chem., Vol. 55, No. 23, 2007 López-Nicolás et al.

not reduce the high variations observed in ∆E* in the absenceof CDs during the first 60 min. Indeed, when maltosyl-�-CDwas added to the medium, these variations were higher thanthose observed for the control. Moreover, the degradation ofinitial color observed in the absence of CD was mostly due todramatic depletions in both lightness (L*) and the blue–yellowchromatism (b*), whereas the greenness–redness (a*) did nothave the same weight in ∆E*.

Bearing in mind that the results obtained for the evolutionof color parameters of banana juice in the presence of CDs areopposite that observed for other fruits juices (37, 40), the nextstep in this investigation was to look for a kinetic model tojustify these differences.

Kinetic Model of the Banana Juice Color in the Presenceand Absence of CD. To clarify the action mechanism of CDsas a compound that can produce a dual effect on enzymaticbrowning (as an activator or inhibitor depending upon the fruitsource), a kinetic model is proposed.

As shown in Figure 1, the addition of CDs to apple, grape,pear, and peach juices inhibited enzymatic browning. The kineticmodel explaining this behavior has been published by our group(37) and is based on the complexation of the different phenolspresent in these fruit juices by CDs, where by only moleculesof substrate PPO may enter a CD molecule (stoichiometry of1:1). In this way, their oxidation to o-quinones and subsequentpolymerization to brown pigments are prevented. The mainphenolic compounds present in these fruit juices that can beoxidized by PPO are shown in Table 3 (48–52). As can beseen in this table, these compounds are of a hydrophobic natureand have been reported as guest molecules for differentCDs (53–57). In the model described for the browning of thesefruit juices in the presence of CDs, referring to 1 h after juicing,the velocity of the process is determined by a Michaelis–Mentenequation, where the free concentration of the substrate isnegligible with respect to the Km.

However, this kinetic model is not suitable for clarifyng theactivating effect of CDs on banana juice browning because,although banana fruit present a high concentration of phenols,the nature of these compounds is very different from thatobserved for the phenols present in the other fruits tested.Several papers published show how dopamine, a hydrophiliccompound, is the main natural phenol present in banana (58, 59).Moreover, the oxidation of dopamine by PPO leads to a reactionproduct (dopaminochrome: λmax ) 480 nm; ε ) 3300 M-1

cm-1), which has a greater influence on the final color of bananajuice than the oxidation of other phenols, with an ε value of

about 1100 M-1 cm-1 (12, 59). However, although the literaturecontains many studies showing how dopamine is oxidized byPPO, our group showed that the enzymatic activity of bananaPPO using dopamine as the substrate was unaffected in thepresence of CDs because this hydrophilic phenol does not forminclusion complexes (38).

For these reasons and to develop the kinetic model proposedin this paper to study the activating effect of CD on bananajuice browning, two hypotheses were assumed: (i) CDs are ableto complex the natural browning inhibitors present in banana;and (ii) the main substrate of PPO, i.e., dopamine, can not becomplexed by CD, and the complexation constant between CDsand natural PPO substrates can be ignored.

The process can be expressed as indicated in Scheme 1, whereE is the enzyme, S is the substrate, CDF is the free cyclodextrin,IF is the free PPO inhibitor, CD–I is the complex between PPOinhibitors and CDs, Kci is the complexation constant betweenCDs and natural PPO inhibitors, and Ki is the inhibition constant.

The main objective of this kinetic model was to calculatethe complexation constant between CDs and the natural PPOinhibitors present in banana juice using our experimental data.For this, we used the evolution of the total color difference(∆E*) data because this value takes into account the differencesin L*, a*, and b* between the samples and standards.

To evaluate the variations of ∆E* when the CD concentrationis increased and to calculate the apparent complexation constantsbetween CDs and the mixture of inhibitors present in bananajuice, a Michaelis–Menten defining equation for linear competi-tive inhibition must be obtained as a function of the only knownparameter, that is, the total CD concentration.

Therefore, the velocity of ∆E* evolution (V) can be expressedas

v)Vmax[S]

Km(1+[I]F

Ki)+ [S]

(1)

To develop the kinetic model proposed, we have assumedthat the Km values of some phenolic substrates of PPO frombanana present higher values than the free concentrations ofpolyphenolic compounds in the juice elaborated from thatfruit (60, 61), which means that the concentration of substrate

Table 3. Effect of Adding CDs on the Browning of Several Fruit Juices and the Complexation of Their Main Polyphenolic Compounds

fruit juiceprincipal polyphenolic

compounds PPO substrates reference complexation by CDs referenceeffect of CDon browning

pear, apple, peach, grape

chlorogenic acid + 45 + 50

inhibitorcaffeic acid + 46 + 51catechin + 47 + 52epicatechin + 48 + 53quercetin + 49 + 54

banana dopamine + 12 – 35 activator

Scheme 1. CDs as Activating Agents of Juice Enzymatic Browning Table 4. Values of the First and Second Apparent Kinetic Constants (kapp

and k′app, and Apparent Complexation Constant (Kci) for the Interactionbetween Maltosyl-�-CD and Banana Juice

foodmaltosyl-�-CD

(mM)k′app

(min-1)kapp

(min-1)Kci

(mM-1)

banana juice 0 0.0274 0.098 ( 0.001 27.026 ( 0.21210 0.032330 0.044060 0.054090 0.0607

Activation of Banana Juice Enzymatic Browning J. Agric. Food Chem., Vol. 55, No. 23, 2007 9659

is negligible with respect to the Km, i.e., [S] , Km. In this case,the velocity of ∆E* evolution can be expressed as

v)kapp[S]

(1+[I]F

Ki) (2)

where kapp is a kinetic constant defined as

kapp )Vmax

Km(3)

The next step was to calculate the [I]F to be substituted in eq2 and, therefore, develop the velocity of ∆E*.

For this, the Kci, i.e., the complexation constant between CDsand PPO inhibitors present in banana juice, is defined as

Kci )[I]F[CD]F

[CD–I](4)

The mass balance of the inhibitor and CDs is represented by

[I]T ) [CD–I]+ [I]F (5)

[CD]T ) [CD–I]+ [CD]F (6)

(where both subscripts T and F denote the total and freeconcentrations, respectively).

Taking into account eqs 4–6 and assuming that [I]T , [CD]T,then [CD]T = [CD]F and the [I]F can be expressed as

[I]F )Kci[I]T

[CD]T +Kci(7)

Substituting eq 7 into eq 2, the velocity of ∆E* evolutioncan be expressed as

v)kapp[S]

(1+

Kci[I]T

[CD]T +Kci

Ki)

(8)

The next step was to calculate the complexation constantbetween CDs and the PPO inhibitors present in banana juice,i.e., Kci, using eq 8. To do this, a second apparent kineticconstant k′app, dependent upon the[CD]T, was defined as

k′app )kappKi

Ki +Kci[I]T

Kci + [CD]T

(9)

To calculate the second apparent kinetic constant (k′app) valuesfor different maltosyl-�-CD concentrations, we used the modelproposed by Soliva-Fortuny et al. (62, 63) to study the influenceof enzymatic browning on the color parameters of pears andapples in several conditions. The evolution of the total colordifference (∆E*) experimental data were fitted (- - - in Figure4) to the first-order fractional model described in eq 10 bynonlinear regression

∆Ef*)∆E*+∆Ef · e-k'appt (10)

where ∆E* is the current value of the total color difference,∆Ef* is the nonzero value of the parameter upon prolongedstorage, t is the storage time, and k′app is the second apparentkinetic constant.

The values obtained for k′app at increasing concentrations ofmaltosyl-�-CD are shown in Table 4.

Finally, to calculate the complexation constant between CDsand natural PPO inhibitors (Kci), k′app was plotted versus thetotal CD concentration ([CD]T) (Figure 5). The values of thefirst kinetic constant, kapp, and Kci were determined using SigmaPlot (SPSS, Inc.), fitting the data by regression to a rationalequation (eq 11) of the type

k′app )kappKci + kapp[CD]T

Kci +Kci[I]T

Ki+ [CD]T

(11)

The data are depicted in Table 4. The Kci value obtained forthe complexation of the natural inhibitors of PPO present inbanana juice and maltosyl-�-CD were of the same order ofmagnitude as those described by our group for the complexationof the mixture of phenols present in apple juice and the sametype of CD (37).

CONCLUSIONS

The results presented in this paper show that any antioxidantagent used to avoid enzymatic browning must be testedindividually for each food because an opposite effect may beproduced. Knowledge of the kinetic models of the enzymaticprocesses occurring in foods is essential to understand theirmacroscopic behavior, as in the case of organoleptic properties,such as color. The surprising observation that CDs chosen asnatural antibrowning agents of fruit juice may behave asprobrowning agents, depending upon by the fruit source, isexplained in this investigation through the biochemical modelreflecting the interaction between natural compounds presentin banana juice and CDs. Therefore, the presence of hydrophobicor hydrophilic phenols in the fruit structure and the inability ofCDs to complex dopamine mean that a recognized antibrowningagent, such as CDs, can be converted to a probrowning agent,leading to changes in color parameters not previously shown.Moreover, determination of the apparent complexation constantbetween the natural polyphenoloxidase inhibitors present inbanana juice and CDs is of great importance to potential usersof these food color modulators because, although browning foodis usually associated in a loss of nutritional and sensorial quality,there is a restricted group of foods where darkening is preferable.

Figure 5. Effect of the maltosyl-�-CD concentration on k′app in bananajuice at 25 °C. Each data point is the mean of three replicates.

9660 J. Agric. Food Chem., Vol. 55, No. 23, 2007 López-Nicolás et al.

ABBREVATIONS USED

CD, cyclodextrin; PPO, polyphenol oxidase.

LITERATURE CITED

(1) Ministerio de agricultura pesca y alimentación (MAPA). Consumoalimentario; MAPA: Madrid, Spain, 2006, www.mapa.es.

(2) Jagtiani, J.; Chang, H. T.; Sakai, W. S. Guava. In Tropical FruitProcessing; Academic Press: New York, 1988.

(3) Becker, D. K.; Dugdale, B.; Smith, M. K.; Harding, R. M.; Dale,J. L. Genetic transformation of Cavendish banana (Musa spp.AAA group) cv “Grand Nain” via microprojectile bombardment.Plant Cell Rep. 2000, 19, 229–234.

(4) Ortiz, R.; Vuylsteke, D. Quantitative variation and phenotypiccorrelations in banana and plantain. Sci. Hortic. 1998, 72, 239–253.

(5) Forster, M. P.; Rodríguez, E.; Díaz Romero, C. Differentialcharacteristics in the chemical composition of bananas fromTenerife (Canary Islands) and Ecuador. J. Agric. Food Chem.2002, 50, 7586–7592.

(6) Cano, M. P.; de Ancos, B.; Matallana, M. C.; Cámara, M.;Reglero, G.; Tabera, J. Differences among Spanish and Latin-American banana cultivars: Morphological, chemical and sensorycharacteristics. Food Chem. 1997, 59, 411–419.

(7) Boudhrioua, N.; Giampaoli, P.; Bonazzi, C. Changes in aromaticcomponents of banana during ripening and air-drying. Lebensm.-Wiss. Technol. 2003, 36, 633–642.

(8) Mayr, D.; Märk, T.; Lindinger, W.; Brevard, H.; Yeretzian, C.Breath-by-breath analysis of banana aroma by proton transferreaction mass spectrometry. Int. J. Mass Spectrom. 2003, 223,743–756.

(9) Sánchez-Ferrer, A.; Rodríguez-López, J. N.; García-Cánovas, F.;García-Carmona, F. Tyrosinase: A comprehensive review of itsmechanism. Biochim. Biophys. Acta 1995, 1247, 1–11.

(10) Cano, M. P.; de Ancos, B.; Gloria Lobo, M.; Santos, M.Improvement of frozen banana (Musa caVendishii, cv. Enana)colour by blanching: Relationship between browning, phenols andpolyphenol oxidase and peroxidase activities. Z. Lebensm.-Unters.Forsch. A 1997, 204, 60–65.

(11) Yang, C. P.; Fujita, S.; Ashrafuzzaman, M. D.; Nakamura, N.;Hayashi, N. Purification and characterization of polyphenoloxidase from banana (Musa sapientum L.) pulp. J. Agric. FoodChem. 2000, 48, 2732–2735.

(12) Sojo, M. M.; Nuñez-Delicado, E.; García-Carmona, F.; Sánchez-Ferrer, A. Partial purification of a banana polyphenol oxidase usingTriton X-114 and PEG 8000 for removal of polyphenols. J. Agric.Food Chem. 1998, 46, 4924–4930.

(13) Yang, C.-P.; Fujita, S.; Kohno, K.; Kusubayashi, A.; Ashrafuzzaman,M. D.; Hayashi, N. Partial purification and characterization ofpolyphenol oxidase from banana (Musa sapientum L.) peel. J.Agric. Food Chem. 2001, 49, 1446–1449.

(14) Umit Unal, M. Properties of polyphenol oxidase from Anamurbanana (Musa caVendishii). Food Chem. 2007, 100, 909–913.

(15) Wuyts, N.; De Waele, D.; Swennen, R. Extraction and partialcharacterization of polyphenol oxidase from banana (Musaacuminata Grande naine) roots. Plant Phys. Biochem. 2006, 44,308–314.

(16) Sapers, G. M.; Hicks, K. B.; Miller, R. L. Antibrowning agents.In Food AdditiVes, 2nd ed.; Branen, A. L., Davidson, P. M.,Salminen, S., Thorngate, J. H., Eds.; Marcel Dekker, Inc.: NewYork, 2001; pp 543–561.

(17) Valero, E.; Varón, R.; García-Carmona, F. A kinetic study ofirreversible enzyme inhibition by an inhibitor that is renderedunstable by the enzymic catalysis: the inhibition of polyphenoloxidase by L-cysteine. Biochem. J. 1991, 277, 869–874.

(18) Valero, E.; Varón, R.; García-Carmona, F. Kinetic study of theeffect of metabisulfite on polyphenol oxidase. J. Agric. FoodChem. 1992, 40, 904–908.

(19) Sapers, G. M. Browning of foods. Control by sulfites, antioxidantsand other means. Food Technol. 1993, 47, 75–84.

(20) Joslyn, M. A.; Braverman, J. B. S. The chemistry and technologyof the pre-treatment and preservation of fruit and vegetablesproducts with sulphur dioxide and sulfites. AdV. Food Res. 1954,5, 97–160.

(21) Wedzicha, B. L.; Bellion, I.; Goddard, S. J. Inhibition of browningby sulfites. AdV. Exp. Med. Biol. 1991, 289, 217–236.

(22) Krapfenbauer, G.; Kinner, M.; Gossinger, M.; Schonlechner, R.;Berghofer, E. Effect of thermal treatment on the quality of cloudyapple juice. J. Agric. Food Chem. 2006, 54, 5453–5460.

(23) Chutintrasri, B.; Noomhorm, A. Thermal inactivation of polyphe-noloxidase in pineapple puree. LWT–Food Sci. Technol. 2006,39, 492–495.

(24) Castro, I.; Macedo, B.; Teixeira, J. A.; Vicente, A. A. The effectof electric field on important food-processing enzymes: Compari-son of inactivation kinetics under conventional and ohmic heating.J. Food Sci. 2004, 69, 696–701.

(25) Giner, J.; Ortega, M.; Mesegué, M.; Gimeno, V.; Barbosa-Cánovas, G. V.; Martín, O. Inactivation of peach polyphenoloxi-dase by exposure to pulsed electric fields. J. Food Sci. 2002, 67,1467–1472.

(26) García-Carmona, F.; Valero, E.; Cabanes, J. Effect of L-prolineof mushroom tyrosinase. Phytochemistry 1988, 27, 1961–1964.

(27) Valero, E.; García-Carmona, F. Hysteresis and cooperativebehavior of a latent plant polyphenoloxidase. Plant Physiol. 1992,98, 774–776.

(28) Valero, E.; Escribano, J.; García-Carmona, F. Reactions of4-methyl-o-benzoquinone, generated chemically or enzymatically,in the presence of L-proline. Phytochemistry 1988, 27, 2055–2061.

(29) Grice, H. C. Food antioxidants: International perspectives, pro-ceedings sponsored by the International Life Sciences. Institute-Nutrition Foundation, Antioxidant Technical Committee. FoodChem. Toxicol. 1986, 24, 819–825.

(30) Shahidi, F.; Janitha, P. K.; Wanasundara, P. D. Phenolic antioxi-dants. Crit. ReV. Food Sci. Nutr. 1992, 32, 67–75.

(31) Lee, M. Inhibitory effect of banana polyphenol oxidase duringripening of banana by onion extract and Maillard reactionproducts. Food Chem. 2007, 102, 146–149.

(32) Billaud, C.; Regaudie, E.; Fayad, N.; Richard-Forget, F.; Nicolas,J. Enzymatic browning and its prevention. In Effect of Cyclodex-trins on Polyphenol Oxidation Catalysed by Apple PolyphenolOxidase; American Chemical Society: Washington, D.C., 1995;pp 295–312.

(33) Hicks, K. B.; Sapers, G. M.; Seib, P. A. Process for preservingraw fruit and vegetables juices using cyclodextrins and composi-tions thereof. U.S. Patent, 4,975,293, 1990.

(34) Saenger, W. Cyclodextrin inclusion compounds in research andindustry. Angew. Chem., Int. Ed. Engl. 1980, 19, 344–362.

(35) López-Nicolás, J. M.; Bru, R.; García-Carmona, F. Kineticcharacteristics of the enzymatic conversion in presence of cyclo-dextrins: Study of the oxidation of polyunsaturated fatty acids bylipoxygenase. Biochim. Biophys. Acta 1997, 1347, 140–150.

(36) López-Nicolas, J. M.; Bru, R.; Garcia-Carmona, F. Enzymaticoxidation of linoleic acid by lipoxygenase forming inclusioncomplexes with cyclodextrins as starch model molecules. J. Agric.Food Chem. 1997, 45, 1144–1148.

(37) López-Nicolás, J. M.; Núñez-Delicado, E.; Sánchez-Ferrer, A.;García-Carmona, F. Kinetic model of apple juice enzymaticbrowning in the presence of cyclodextrins: The use of maltosyl-�-cyclodextrin as secondary antioxidant. Food Chem. 2007, 101,1164–1171.

(38) Sojo, M. M.; Núñez-Delicado, E.; García-Carmona, F.; Sánchez-Ferrer, A. Cyclodextrins as activator and inhibitor of banana pulppolyphenol oxidase. J. Agric. Food Chem. 1999, 47, 518–523.

(39) Meléndez-Martínez, A. J.; Vicario, I. M.; Heredia, F. J. Applicationof tristimulus colorimetry to estimate the carotenoids content inultrafrozen orange juices. J. Agric. Food Chem. 2003, 51, 7266–7270.

(40) Nuñez-Delicado, E.; Serrano-Megias, M.; Pérez-López, A. J.;López-Nicolás, J. M. Polyphenol oxidase from Dominga tablegrape. J. Agric. Food Chem. 2005, 53, 6087–6093.

Activation of Banana Juice Enzymatic Browning J. Agric. Food Chem., Vol. 55, No. 23, 2007 9661

(41) López-Nicolás, J. M.; Núñez-Delicado, E.; Pérez-López, A. J.;Carbonell Barrachina, A.; Cuadra-Crespo, P. Determination ofstoichiometric coefficients and apparent formation constants for�-cyclodextrin complexes of trans-resveratrol using reversed-phase liquid chromatography. J. Chromatogr., A 2006, 1135, 158–165.

(42) Fujimura, K.; Ueda, T.; Kitagawa, M.; Takayanagi, H.; Ando, T.Reversed-phase retention behavior of aromatic compounds involv-ing �-cyclodextrin inclusion complex formation in the mobilephase. Anal. Chem. 1986, 58, 2668–2674.

(43) Cheng, G. W.; Crisosto, C. H. Browning potential, phenoliccomposition, and polyphenoloxidase activity of buffer extractsof peach and nectarine skin tissue. J. Am. Soc. Hortic. Sci. 1995,120, 835–838.

(44) López-Nicolás, J. M.; Bru, R.; Sanchez-Ferrer, A.; Garcia-Carmona, F. Use of “soluble lipids” for biochemical processes:Linoleic acid–cyclodextrin inclusion complexes in aqueous solu-tions. Biochem. J. 1995, 308, 151–154.

(45) Martin Del Valle, E. M. Cyclodextrins and their uses: A review.Process Biochem. 2004, 39, 1033–1046.

(46) Szente, L.; Szejtli, J. Cyclodextrins as food ingredients. TrendsFood Sci. Technol. 2004, 15, 137–142.

(47) Veiga, M. D.; Merino, M.; Fernandez, D.; Lozano, R. Charac-terization of some cyclodextrin derivatives by thermal analysis.J. Therm. Anal. Calorim. 2002, 68, 511–516.

(48) Nkya, E.; Kouno, C.; Li, Y. J.; Yang, C. P.; Hayashi, N.; Fujita,S. Purification and characterization of polyphenol oxidase fromgarland chrysanthemum (Chrysanthemum coronarium L.). J.Agric. Food Chem. 2003, 51, 5467–5471.

(49) Carmona, F. G.; Pedreño, E.; Galindo, J. D.; Cánovas, F. G. Anew spectrophotometric method for the determination of cresolaseactivity of epidermis tyrosinase. Anal. Biochem. 1979, 95, 433–435.

(50) Jimenez-Atienzar, M.; Cabanes, J.; Gandia-Herrero, F.; Garcia-Carmona, F. Kinetic analysis of catechin oxidation by polyphenoloxidase at neutral pH. Biochem. Biophys. Res. Commun. 2004,319, 902–910.

(51) Heimdal, H.; Bro, R.; Larsen, L. M.; Poll, L. Prediction ofpolyphenol oxidase activity in model solutions containing variouscombinations of chlorogenic acid, (–)-epicatechin, O2, CO2,temperature, and pH by multiway data analysis. J. Agric. FoodChem. 1997, 45, 2399–2406.

(52) Jimenez, M.; Garcia-Carmona, F. Oxidation of the flavonolquercetin by polyphenol oxidase. J. Agric. Food Chem. 1999, 47,56–60.

(53) Irwin, P. L.; Pfeffer, P. E.; Doner, L. W.; Sapers, G. M.; Brewster,J. D.; Nagahashi, G.; Hicks, K. B. Binding geometry, stoichiom-etry, and thermodynamics of cyclomalto-oligosaccharide (cyclo-dextrin) inclusion complex formation with chlorogenic acid, the

major substrate of apple polyphenol oxidase. Carbohydr. Res.1994, 256, 13–27.

(54) Rodrigues, E.; Vaz1, S.; Gil, V. M. S. S.; Caldeira, M.; Moreirada Silva, A. M. G. Inclusion of polyphenol oxidase substrates in�-cyclodextrin: A 1H-NMR study. J. Inclusion Phenom. Macro-cyclic Chem. 200244, 395–397.

(55) Jullian, C.; Miranda, S.; Zapata-Torres, G.; Mendizabal, F.; Olea-Azar, C. Studies of inclusion complexes of natural and modifiedcyclodextrin with (+)-catechin by NMR and molecular modeling.Bioorg. Med. Chem. 2007, 15, 3217–24.

(56) Ishizu, T.; Kintsu, K.; Yamamoto, H. NMR study of the solutionstructures of the inclusion complexes of �-cyclodextrin with (+)-catechin and (–)-epicatechin. J. Phys. Chem. B 1999, 103, 8992–8997.

(57) Pralhad, T.; Rajendrakumar, K. Study of freeze-dried quercetin–cyclodextrin binary systems by DSC, FT-IR, X-ray diffraction,and SEM analysis. J. Pharm. Biomed. Anal. 2004, 34, 333–339.

(58) Kanazawa, K.; Sakakibara, H. High content of dopamine, a strongantioxidant, in Cavendish banana. J. Agric. Food Chem. 2000,48, 844–848.

(59) Sojo, M. M.; Nunez-Delicado, E.; Sanchez-Ferrer, A.; Garcia-Carmona, F. Oxidation of salsolinol by banana pulp polyphenoloxidase and its kinetic synergism with dopamine. J. Agric. FoodChem. 2000, 48, 5543–5547.

(60) Fayad, N.; Marchal, L.; Billaud, C.; Nicolas, J. Comparison of�-cyclodextrin effect on polyphenol oxidation catalysed by purifiedpolyphenol oxidase from different sources. J. Agric. Food Chem.1997, 45, 2442–2446.

(61) van der Sluis, A. A.; Dekker, M.; Skrede, G.; Jongen, W. M. F.Activity and concentration of polyphenolic antioxidants in applejuice. Effect of existing production methods. J. Agric. Food Chem.2002, 50, 7211–7219.

(62) Soliva-Fortuny, R. C.; Biosca-Biosca, M.; Grigelmo, M.; Martín-Belloso, O. Browning, polyphenol oxidase activity and headspacegas composition during storage of minimally processed pears usingmodified atmosphere packaging. J. Sci. Food Agric. 2002, 82,1490–1496.

(63) Soliva-Fortuny, R. C.; Grigelmo-Miguel, N.; Odriozola-Serrano,I.; Gorinstein, S.; Martín-Belloso, O. Browning evaluation ofready-to-eat apples as affected by modified atmosphere packaging.J. Agric. Food Chem. 2001, 49, 3685–3690.

Received for review May 8, 2007. Revised manuscript receivedSeptember 5, 2007. Accepted September 19, 2007. This work wassupported by AGL2007-65907 (MEC, FEDER, Spain). J.M.L.N. holdsa contract with the “Programa Ramón y Cajal” (FEDER, MEC, Spain).

JF0713399

9662 J. Agric. Food Chem., Vol. 55, No. 23, 2007 López-Nicolás et al.